Action

Document classifier

Tag each file by type

Document classifier reads each incoming file and labels what kind of document it is (bank statement, application, invoice, ID), so later steps can route or bundle by type.

How it works

When a submission arrives as a stack of files jumbled together, this node is the sorter. It looks at each file and tags it with a type so the rest of your flow knows a bank statement from a driver's license from a signed application, without a person opening every PDF to check.

This matters because most later steps only make sense on the right kind of document. You want to run extraction on the bank statements, file the ID away, and route the application to a reviewer. Document classifier puts a label on each file so those steps can pick out exactly what they need.

Think of it as the intake clerk who stamps each page before it goes anywhere. Once files are tagged, you can send each type down its own path or gather all files of one type together for the next step.

Common use cases

Real situations where this node earns its place in a flow.

Sort a mixed pile

When an applicant uploads statements, an ID, and a contract together, tag each file so the right step handles it.

Route by document type

Recognize what each file is and branch the flow so invoices and statements go different ways.

Drop what you do not need

Identify and skip irrelevant attachments before they cost you an extraction.

Document classifier in a workflow

A simple example of where this node sits in a flow.

Direct upload Web /jobs/new form Document classifier Tag each file by type Extract PDFs → structured data

What you set up

  • Choose the document types you expect to receive and want labeled.
  • Decide what happens to each type next (route it, bundle it, or set it aside).
  • Handle the catch-all: pick where files that match no known type should go.

Common questions

What happens to a file it can't recognize?

It gets an unknown or catch-all label rather than being dropped, so you can route those files for a person to look at instead of losing them silently.

Is this the same as Classify?

No. Classify (scrub) branches the whole job on rules you write about its values. Document classifier works file by file, labeling each one by what kind of document it is.